On the Prosodic Coding of Focus in Paraguayan Guaraní

An utterance conveys meaning not just through the words that are uttered (and the way these words combine to form larger expressions), but also through its prosody, i.e. its rhythm and tune. The contribution of prosody to the truth-conditional meaning of an utterance as well as to its informationstructural meaning, i.e. the way in which the utterance is used in discourse, is determined at the interface between the intonational phonology of the language, its syntax, and its semantics. The contribution of prosody to meaning has been studied extensively for a number of languages, including English (e.g., Bolinger 1958, Jackendoff 1972, Pierrehumbert 1980), Japanese (Beckman & Pierrehumbert 1986), Italian (D’Imperio 2002), Spanish (Face 2001), and Mandarin (Hu 2002). One insight of this research has been that there is “extensive cross-linguistic variation in the phonology of prosody” (Beaver & Clark 2008: 10, see also e.g., Jun 2005 and Ladd 1996). However, because the interface between intonational phonology and a theory of meaning has been explored mainly for English and German (e.g., Pierrehumbert & Hirschberg 1990, Rooth 1992, Büring 1997, Schwarzschild 1999, Féry & Samek-Lodovici 2006), the questions of how cross-linguistic differences in intonational phonology are reflected semantically, and whether the same (or similar) principles of interpretation govern the semantics of prosody across languages remain largely unaddressed. The current study represents the first part of a larger project examining the meaning of prosody in Paraguayan Guaraní, a Tupí-Guaraní language spoken by approximately four million people in Paraguay and surrounding countries. Our long-term goals are to determine the meaningful units of Paraguayan Guaraní (henceforth Guaraní) prosodic structure and how the prosody of a Guaraní utterance contributes to its meaning. To begin to address these goals, the current study involves a production experiment designed to explore the prosody of simple Guaraní utterances realized in three focus conditions. We define the focus of an utterance as that part of the utterance that answers the Question under Discussion (QUD, cf., Ginzburg 1996, Roberts 1996). Thus, we use the term focus to denote a semantic/pragmatic property, not a prosodic property. The QUD may be explicitly given, as in the examples in (1), or implicit in the discourse context. The three focus conditions included in the current study are illustrated in (1) for the English utterance Maria is writing. In (1a), which illustrates the ‘verb new focus’ condition, the explicit QUD is What is Maria doing?. Since the verb phrase is writing of the utterance Maria is writing answers the QUD, it is a focus of the utterance; since the verb phrase is not mentioned in the QUD (or in the preceding discourse), we refer to it as a ‘new focus’. The denotation of Maria in the answer to this QUD is given, or discourse-old (cf., Prince 1992), and not a focus. In (1b), which illustrates the ‘verb contrastive focus’ condition, the verb phrase is writing is again a focus of the answer, but the verb phrase is now what we call contrastively focused because the QUD inquires about which of the alternatives explicitly given in the QUD is the one that answers the QUD. The denotation of the subject Maria is again given and not a focus. In the ‘subject contrastive focus’ condition illustrated in (1c), the QUD inquires about which of the explicitly given alternatives,